{
  "docId": "019dd923-5e88-73ef-bd5d-06b04d219fea",
  "docSlug": "dd91c78f6570bf29",
  "documentTitle": "2023 Air Street Capital The State of AI Report 2023",
  "authorId": "AirStreetCapital",
  "authorName": "Air Street Capital",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "vc_research",
  "sourceTypeLabel": "VC research",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.777,
  "pageNumber": 54,
  "pageCount": 163,
  "prevPage": 53,
  "nextPage": 55,
  "slideType": "industry_trends",
  "function": "summarize",
  "density": "dense",
  "nDataPoints": 0,
  "notes": "The slide highlights two specific research papers/models in the field of computer vision.",
  "elementsJson": [
    "headline_text",
    "paragraph",
    "bullet_list",
    "photo"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-06b04d219fea/54",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea#slide-54",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Zero-shot depth models have recently been used as conditioning for better image generation.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a224-dcbd34a7f82e",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.2,
        "w": 0.56,
        "x": 0.22,
        "y": 0.72
      },
      "kind": "image",
      "text": "Visual examples of depth estimation results",
      "attrs": null,
      "subkind": "infographic",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "de06b3f0-41e1-4aed-83c1-b7eca3e15b64",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.15,
        "w": 0.9,
        "x": 0.05,
        "y": 0.35
      },
      "kind": "list",
      "text": "ZeroDepth: Towards Zero-Shot Scale-Aware Monocular Depth Estimation is able to predict metric depth for images from different domains and different camera parameters. They jointly encode image features and camera parameters which enables the network to reason over the size of objects and train in a variational framework. The depth network ends up learning 'scale priors' that can be transferred across datasets.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "5047ae15-90b2-492a-899d-d239f53d3d0f",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.1,
        "w": 0.9,
        "x": 0.05,
        "y": 0.55
      },
      "kind": "list",
      "text": "ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth is a relative depth model with an additional module fine-tuned on metric depth. This is the first model to train on multiple datasets without a significant drop in performance and able to generalise across both indoor and outdoor domains.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "d025f37c-5c0c-4620-aae6-6534d8d4709c",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.1,
        "w": 0.9,
        "x": 0.05,
        "y": 0.21
      },
      "kind": "paragraph",
      "text": "Zero-shot depth models have recently been used as conditioning for better image generation. This only requires relative depth prediction, while other downstream applications such as robotics require metric depth which so far has not generalised well across datasets.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "5a1b48f0-3f04-44dd-ba0e-55850fe38e49",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.35,
        "x": 0.05,
        "y": 0.13
      },
      "kind": "title",
      "text": "Zero-shot metric depth is here",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "d41f2eb6-6a2f-402e-91ad-92dd4c7a7146",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Before-After-Bridge",
      "slug": "before-after-bridge",
      "agent": "Storyteller",
      "layer": "slide",
      "matchId": "c93f11ae-11d2-4d13-95e0-2f9d61c590be",
      "evidence": "Visual examples of depth estimation results",
      "confidence": 0.6
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 120,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-41afd44ef59f",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Research + Industry sections inventory model, compute, funding facts.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 120,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-523bfb7f96e6",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Escalating capabilities, compute concentration and capital flows.",
      "position": 2,
      "confidence": 45,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 58,
      "from": 50,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-f70f6435b722",
      "evidence": "Text-to-video, instruction editing, Gaussian splatting, NeRFs, depth, SAM, DINOv2, weather, music.",
      "position": 6,
      "objective": "Generative media and visual foundation models proliferate",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 72,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}